- Email: [email protected]
Genetics of social behaviour in fire ants
Research Update
TRENDS in Genetics Vol.18 No.5 May 2002
221
Research News
Genetics of social behaviour in fire ants Andrew F.G. Bourke A recent study is the first to sequence a gene known to underlie a complex social phenotype. In the fire ant, Solenopsis invicta, a single allelic difference at the Gp-9 locus specifies the number of queens a colony has, and hence the social structure of the colony. Gp-9 appears to encode a protein implicated in chemical recognition of nestmates, consistent with workers determining queen number by selectively executing queens as a function of workers’ and queens’ Gp-9 genotypes. Other Solenopsis species exhibit the same social and genetic polymorphism. This study pioneers the integrated understanding of the evolution of social behaviour at molecular, individual and social levels.
For decades, evolutionary theorists have sought to understand the conditions for social evolution by imagining the fate of genes underlying social behaviour. Modern attempts started in 1963 with W.D. Hamilton’s kin selection theory [1], which proposed that altruism evolves by the spread of genes whose bearers direct benefits to relatives. Modelling social evolution requires invoking genes, even imaginary ones. For biological evolution to occur, genes underpinning the trait of interest must exist. Moreover, although still controversial, kin selection is now supported strongly by a large body of evidence (e.g. Refs [2,3]). But to two sets of onlookers, this carefree invocation of genes for social behaviour has remained an irritant. Some ethologists suspect that complex behaviours are unlikely to be so tightly influenced by genes. Molecular geneticists fret that nothing is known about the identity or products of such genes. An outstanding new study by Michael Krieger and Ken Ross (Dept of Entomology, University of Georgia, USA) should go far in satisfying both sets of critics [4]. Krieger and Ross [4] have sequenced a gene known to underlie the expression of a complex social polymorphism in the fire ant, Solenopsis invicta. They have found that eight nucleotide differences in the coding sequences of the two alleles at this locus determine, through effects on http://tig.trends.com
worker behaviour, whether a colony has one or several queens. The work therefore establishes convincingly that the presence or absence of a multifaceted social phenotype can depend on differences at a single locus. Krieger and Ross followed several additional lines of investigation, each of interest to molecular geneticists and novel in the study of social evolution. These include the tentative identification of the gene product, an examination of the strength of selection on the gene, and its phylogenetic analysis. Gyny genetics in fire ants
The fire ant S. invicta is a native of South America that has become a notorious pest in the USA since its introduction in the 1930s. The first populations of S. invicta discovered in the USA were monogynous and multicolonial; that is, a single queen heads each colony, and her sterile daughters (the workers) are hostile to individuals from other colonies. However, from the 1970s onwards, populations were increasingly found that were polygynous and unicolonial [5]. Their colonies house many queens, including those adopted from outside the nest. The selective reasons for the evolution of unicolonial polygyny in S. invicta are not settled, but could stem from the greater ecological success of this social organization as population density increases [5,6]. Nonetheless, variation in gyny (queen number) clearly represents a major social polymorphism in introduced fire ant populations.
Over the past ten years, a series of painstaking studies by Ken Ross, Laurent Keller and co-workers, has established and unravelled the genetic basis of gyny variation in S. invicta (e.g. Refs [7–12]). This variation is controlled by a locus, Gp-9, whose two alleles (B and b) have several effects on development, worker behaviour and queen phenotype (Table 1). In particular, BB workers in the absence of Bb workers do not tolerate multiple queens in their colonies, so monogynous colonies are always headed by a BB queen. By contrast, Bb workers tolerate multiple queens, but only if these queens bear the b allele. Heterozygote workers somehow detect BB queens and kill them by biting, so polygynous colonies are always headed by Bb queens. Experiments suggest that Bb workers detect BB queens by sensing the lack of a surface chemical cue associated with the b allele, because workers rubbed against BB queens were killed by other workers, but those rubbed against Bb queens were not [13]. The b allele therefore acts as a ‘green beard’ gene [13,14]. Its bearers (Bb workers) recognize and discriminate against nonbearers (BB queens) on the basis of an external label (a ‘green beard’), and as a result direct benefits to co-bearers. Through the net outcome of these complex effects, and especially through the influence on worker behaviour, the presence or absence of the b allele among workers specifies the social structure of the colony (Table 1).
Table 1. Genetic and phenotypic traits of monogynous and polygynous colonies in introduced Solenopsis invicta fire antsa Monogynous colony Queen number One Genotype of queen(s) BB heading colony Queen phenotype High fat reserves, rapid oogenesis (advantageous under monogyny) Worker phenotype BB workers in absence of Bb workers are intolerant of multiple queens
aSee
Polygynous colony Many Bb Low fat reserves, slow oogenesis
Bb workers tolerate multiple queens, but only if they bear the b allele; Bb workers detect and kill BB queens; all bb queens (and workers) die early because b is a recessive lethal
for example Refs [7,8,11].
0168-9525/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S0168-9525(02)02655-0
222
Research Update
Sequencing the Gp-9 gene
Polymorphism in Gp-9 was first detected by protein electrophoresis. By separating the Gp-9 protein electrophoretically and designing degenerate primers on the basis of its amino acid sequence, Krieger and Ross [4] have now isolated and sequenced the Gp-9 gene. The gene turns out to be 1700 bp and to contain five exons and four introns. GenBank BLASTX searches showed its closest match to be genes encoding pheromone-binding proteins in moths. Because these proteins are used in the chemical recognition of conspecific individuals, this is satisfyingly consistent with the presumed effect of Gp-9 in allowing ‘green beard’ recognition of allelic co-bearers. By sequencing the Gp-9 gene from queens of both social forms, Krieger and Ross also determined the nature of allelic variation in the gene: the two alleles differ by nine nucleotide substitutions, eight of which are nonsynonymous; that is, leading to amino acid differences in the alleles’ protein products [4]. Phylogenetics of Gp-9
In its native South America, S. invicta also occurs in polygynous populations, but they do not appear to be unicolonial [15]. Krieger and Ross [4] found that S. invicta queens from polygynous colonies from Argentina always carried a b allele, although in two variants showing slight sequence differences from the b allele in US populations. By sequencing Gp-9 loci in nine additional Solenopsis species and reconstructing their phylogeny, the investigators found a remarkable pattern: obligately monogynous species in the fire ant group possess only B-like alleles and are basal (ancestral to the other groups), whereas nearly all species known to be socially polymorphic had b-like alleles in the polygynous colonies. So, Krieger and Ross suggest that the b allele arose in an ancestrally monogynous BB population and has persisted during subsequent, multiple speciation events, always retaining its influence on social phenotype. Analysis of the ratio of nonsynonymous to synonymous nucleotide substitutions within lineages indicated that the divergence of b-like alleles from B-like alleles has been driven repeatedly by positive selection. Conclusions and prospects
A major achievement of this work is, as Krieger and Ross [4] stress, its http://tig.trends.com
TRENDS in Genetics Vol.18 No.5 May 2002
confirmation that a single gene with a large effect can underlie complex social behaviour. This is contrary to the intuitive view that complex behaviour is influenced by many genes with small effects (but see below). Whether the involvement of genes with major effects will prove exceptional or common in social evolution remains unclear (c.f. Ref. [16]). Another key achievement is pioneering the integrated study of social evolution across all levels of the biological hierarchy. We are now much closer to a complete understanding of a gene for social behaviour, from its sequence identity and protein product, through to its subsequent effects on individual behaviour, through to the summed outcome for the colony’s social structure and the gene’s fitness under natural selection. For this reason, Krieger and Ross’s work marks a crucial stage in the maturation of the study of social evolution as a scientific discipline. In future, it will be important to achieve a similarly comprehensive understanding of comparable genes in other social contexts and taxa. In this respect, note that Gp-9 is probably unusual because it is not a strictly kinselected gene (workers with the b allele favour b-bearing queens irrespective of kinship), and because it has remained polymorphic through b being recessive lethal [13]. The challenge remains to subject kin-selected genes to investigations such as that of Krieger and Ross. This could involve examining facultatively social Hymenoptera, whose populations contain both solitary and social nests (e.g. Ref. [17]). In principle, a single-gene difference could underlie the difference between these two social phenotypes (e.g. presence or absence of a gene for offspring dispersal [2]). However, this phenotypic difference could also stem from conditional expression of alreadyfixed alleles [18]. In theory, a kin-selected gene for helping behaviour retains conditional expression of helper and non-helper phenotypes even at fixation (e.g. Ref. [19]). The next comparable studies could, therefore, be ones where, as in S. invicta, genetic polymorphism is maintained by special circumstances, or conditional expression seems unlikely, or both. A candidate is the heritable ‘anarchic syndrome’ in honey bees, whereby a rare subset of colonies with queens routinely contain reproductive workers [20]. However, researchers will
rarely have the head-start of knowing that the expression of a social trait is correlated with a protein polymorphism. Other remaining questions concern the evolution of polygyny in S. invicta. Why is the b allele associated with multicolonial polygyny in South America but unicolonial polygyny in North America? Why, given the b allele’s presence in South American populations, did polygyny take 40 years to appear in the USA? Does the b allele bring about polygyny as a social novelty, or does it secondarily invade polygynous systems present for ecological reasons? In establishing the feasibility of tackling all these issues, the superb study of Krieger and Ross [4] has set a new standard in genetic research on social evolution. Acknowledgements
I thank Michael Krieger and Tracey Chapman for helpful comments. References 1 Hamilton, W.D. (1963) The evolution of altruistic behavior. Am. Nat. 97, 354–356 2 Bourke, A.F.G. and Franks, N.R. (1995) Social Evolution in Ants, Princeton University Press 3 Crozier, R.H. and Pamilo, P. (1996) Evolution of Social Insect Colonies: Sex Allocation and Kin Selection, Oxford University Press 4 Krieger, M.J.B. and Ross, K.G. (2002) Identification of a major gene regulating complex social behavior. Science 295, 328–332 5 Ross, K.G. and Keller, L. (1995) Ecology and evolution of social organization: insights from fire ants and other highly eusocial insects. Annu. Rev. Ecol. Syst. 26, 631–656 6 Chapman, R.E. and Bourke, A.F.G. (2001) The influence of sociality on the conservation biology of social insects. Ecol. Lett. 4, 650–662 7 Ross, K.G. (1992) Strong selection on a gene that influences reproductive competition in a social insect. Nature 355, 347–349 8 Keller, L. and Ross, K.G. (1993) Phenotypic basis of reproductive success in a social insect: genetic and social determinants. Science 260, 1107–1110 9 Ross, K.G. et al. (1996) Simple genetic basis for important social traits in the fire ant Solenopsis invicta. Evolution 50, 2387–2399 10 Ross, K.G. (1997) Multilocus evolution in fire ants: effects of selection, gene flow and recombination. Genetics 145, 961–974 11 Ross, K.G. and Keller, L. (1998) Genetic control of social organization in an ant. Proc. Natl. Acad. Sci. U. S. A. 95, 14232–14237 12 Keller, L. and Ross, K.G. (1999) Major gene effects on phenotype and fitness: the relative roles of Pgm-3 and Gp-9 in introduced populations of the fire ant Solenopsis invicta. J. Evol. Biol. 12, 672–680 13 Keller, L. and Ross, K.G. (1998) Selfish genes: a green beard in the red fire ant. Nature 394, 573–575 14 Dawkins, R. (1976) The Selfish Gene, Oxford University Press
Research Update
15 Ross, K.G. et al. (1996) Social evolution in a new environment: the case of introduced fire ants. Proc. Natl. Acad. Sci. U. S. A. 93, 3021–3025 16 Orr, H.A. and Coyne, J.A. (1992) The genetics of adaptation: a reassessment. Am. Nat. 140, 725–742 17 Michener, C.D. (1985) From solitary to eusocial: need there be a series of intervening
TRENDS in Genetics Vol.18 No.5 May 2002
species? In Experimental Behavioral Ecology and Sociobiology (Hölldobler, B. and Lindauer, M., eds), pp. 293–305, Gustav Fischer Verlag 18 West-Eberhard, M.J. (1989) Phenotypic plasticity and the origins of diversity. Annu. Rev. Ecol. Syst. 20, 249–278 19 Parker, G.A. (1989) Hamilton’s rule and conditionality. Ethol. Ecol. Evol. 1, 195–211
223
20 Barron, A.B. et al. (2001) Worker reproduction in honey-bees (Apis) and the anarchic syndrome: a review. Behav. Ecol. Sociobiol. 50, 199–208
Andrew F.G. Bourke Institute of Zoology, Zoological Society of London, Regent’s Park, London, UK NW1 4RY e-mail: [email protected]
Antisense RNAs everywhere? E. Gerhart H. Wagner and Klas Flärdh In recent years, systematic searches of both prokaryote and eukaryote genomes have identified a staggering number of small RNAs, the biological functions of which remain unknown. Small RNA-based regulators are well known from bacterial plasmids. They act on target RNAs by sequence complementarity; that is, they are antisense RNAs. Recent findings suggest that many of the novel orphan RNAs encoded by bacterial and eukaryotic chromosomes might also belong to a ubiquitous, heterogeneous class of antisense regulators of gene expression.
Antisense RNAs can be powerful tools to downregulate, at the post-transcriptional level, the expression of targeted genes. In a recent report, a conceptually simple antisense approach was shown to work remarkably well in the Gram-positive bacterium Staphyloccocus aureus [1]. These authors shotgun-cloned genomic segments and generated antisense RNAs from an inducible promoter on the plasmid. Many transformants displayed lethal phenotypes, or severe growth retardation, upon induction. This identified many essential genes that were previously unknown. In Escherichia coli, artificial antisense RNA strategies have been used for many years [2]. Unfortunately, the efficiency of introduced antisense RNA genes in these Gram-negative bacteria varies greatly, and attempts to improve the efficiency by rational design have been rather disappointing [3]. Currently, specific antisense RNA efficiencies cannot be compared between bacterial systems, because intracellular concentrations have generally not been determined. Nevertheless, S. aureus appears to be superior to E. coli in supporting antisense RNA-mediated downregulation of gene expression. Whether this is owing to http://tig.trends.com
differences in intracellular conditions or the presence or absence of certain enzymatic activities is as yet unkown. Antisense RNA and co-suppression (sense RNA) have also been used successfully in plants. Current wisdom suggests that both approaches are related to the natural process of dsRNA-mediated interference (RNAi) [4]. In light of the justified excitement over the successful use of antisense RNA-based approaches, it is surprising that natural antisense RNAs receive little attention. After all, these RNAs do it for a living and might tell us about optimization principles. Antisense interactions involved in biological functions
In Nature, many different biological activities work through an antisense principle. In eukaryotes, processes such as splicing, RNA editing, rRNA modification, and developmental regulation rely on base-pairing between complementary RNAs (or stretches thereof). Small RNAs are the key elements: spliceosomal snRNAs, gRNAs, snoRNAs and stRNAs (see Glossary for definition of RNAs). The recent discovery of RNAi, initially established in Caenorhabditis elegans,
adds to this list [5]. In RNAi, dsRNA is cleaved by the enzyme Dicer, into ~21-nucleotide (nt) dsRNAs called small interfering RNAs (siRNAs). Mediated by an enzyme complex, RISC, target RNA is bound by the complementary RNA – one of the two strands in the short duplex – and subsequently degraded [4]. The antisense principle is crucial for the specificity of target recognition and silencing, and in this context presumably acts to defend against viruses and transposition events. Antisense RNAs as regulators
Historically, Jacob and Monod [6] first suggested that gene regulation in bacteria could be carried out by RNAs (or proteins). About 20 years later, small untranslated antisense RNAs were discovered in bacterial plasmids and shown to regulate the copy numbers of two plasmids, ColE1 [7] and R1 [8]. RNAI (of ColE1; 108 nt) and CopA (of R1; 93 nt) both bind rapidly and irreversibly to their respective target RNAs to block primer RNA maturation or inhibit Rep protein translation, respectively (Fig. 1). Regulation occurs by negative feedback. Numerous studies have given a detailed understanding of both the biology and biochemistry of these systems (for reviews, see Refs [9–11]).
Glossary of RNA terms dsRNA: Double-stranded RNA. gRNA: Short guide RNAs that bind to complementary pre-mRNAs to specify editing. miRNA: Micro RNAs, a class of 22- to 25-nucleotide short RNAs in animals. Functions are as yet unknown, but might be stRNAs. ncRNA: Noncoding RNA. In general, these are non-messenger RNAs. Riboregulator: Regulatory RNA (term used in bacteria). RNAi: dsRNA-mediated RNA interference. Gene silencing using introduced dsRNA with homology to the target RNA. sRNA: Small noncoding RNA. A class of putative riboregulators. siRNA: Short, double-stranded ~22mers that are generated during RNAi; siRNAs mediate subsequent target RNA degradation. snRNA: Heterogeneous class of small nuclear RNAs; for example those involved in splicing. snoRNA: Small nucleolar RNA. Directs rRNA modification. stRNA: Small temporal RNA such as lin-4 and let-7, involved in developmental control.
0168-9525/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S0168-9525(02)02658-6
TRENDS in Genetics Vol.18 No.5 May 2002
221
Research News
Genetics of social behaviour in fire ants Andrew F.G. Bourke A recent study is the first to sequence a gene known to underlie a complex social phenotype. In the fire ant, Solenopsis invicta, a single allelic difference at the Gp-9 locus specifies the number of queens a colony has, and hence the social structure of the colony. Gp-9 appears to encode a protein implicated in chemical recognition of nestmates, consistent with workers determining queen number by selectively executing queens as a function of workers’ and queens’ Gp-9 genotypes. Other Solenopsis species exhibit the same social and genetic polymorphism. This study pioneers the integrated understanding of the evolution of social behaviour at molecular, individual and social levels.
For decades, evolutionary theorists have sought to understand the conditions for social evolution by imagining the fate of genes underlying social behaviour. Modern attempts started in 1963 with W.D. Hamilton’s kin selection theory [1], which proposed that altruism evolves by the spread of genes whose bearers direct benefits to relatives. Modelling social evolution requires invoking genes, even imaginary ones. For biological evolution to occur, genes underpinning the trait of interest must exist. Moreover, although still controversial, kin selection is now supported strongly by a large body of evidence (e.g. Refs [2,3]). But to two sets of onlookers, this carefree invocation of genes for social behaviour has remained an irritant. Some ethologists suspect that complex behaviours are unlikely to be so tightly influenced by genes. Molecular geneticists fret that nothing is known about the identity or products of such genes. An outstanding new study by Michael Krieger and Ken Ross (Dept of Entomology, University of Georgia, USA) should go far in satisfying both sets of critics [4]. Krieger and Ross [4] have sequenced a gene known to underlie the expression of a complex social polymorphism in the fire ant, Solenopsis invicta. They have found that eight nucleotide differences in the coding sequences of the two alleles at this locus determine, through effects on http://tig.trends.com
worker behaviour, whether a colony has one or several queens. The work therefore establishes convincingly that the presence or absence of a multifaceted social phenotype can depend on differences at a single locus. Krieger and Ross followed several additional lines of investigation, each of interest to molecular geneticists and novel in the study of social evolution. These include the tentative identification of the gene product, an examination of the strength of selection on the gene, and its phylogenetic analysis. Gyny genetics in fire ants
The fire ant S. invicta is a native of South America that has become a notorious pest in the USA since its introduction in the 1930s. The first populations of S. invicta discovered in the USA were monogynous and multicolonial; that is, a single queen heads each colony, and her sterile daughters (the workers) are hostile to individuals from other colonies. However, from the 1970s onwards, populations were increasingly found that were polygynous and unicolonial [5]. Their colonies house many queens, including those adopted from outside the nest. The selective reasons for the evolution of unicolonial polygyny in S. invicta are not settled, but could stem from the greater ecological success of this social organization as population density increases [5,6]. Nonetheless, variation in gyny (queen number) clearly represents a major social polymorphism in introduced fire ant populations.
Over the past ten years, a series of painstaking studies by Ken Ross, Laurent Keller and co-workers, has established and unravelled the genetic basis of gyny variation in S. invicta (e.g. Refs [7–12]). This variation is controlled by a locus, Gp-9, whose two alleles (B and b) have several effects on development, worker behaviour and queen phenotype (Table 1). In particular, BB workers in the absence of Bb workers do not tolerate multiple queens in their colonies, so monogynous colonies are always headed by a BB queen. By contrast, Bb workers tolerate multiple queens, but only if these queens bear the b allele. Heterozygote workers somehow detect BB queens and kill them by biting, so polygynous colonies are always headed by Bb queens. Experiments suggest that Bb workers detect BB queens by sensing the lack of a surface chemical cue associated with the b allele, because workers rubbed against BB queens were killed by other workers, but those rubbed against Bb queens were not [13]. The b allele therefore acts as a ‘green beard’ gene [13,14]. Its bearers (Bb workers) recognize and discriminate against nonbearers (BB queens) on the basis of an external label (a ‘green beard’), and as a result direct benefits to co-bearers. Through the net outcome of these complex effects, and especially through the influence on worker behaviour, the presence or absence of the b allele among workers specifies the social structure of the colony (Table 1).
Table 1. Genetic and phenotypic traits of monogynous and polygynous colonies in introduced Solenopsis invicta fire antsa Monogynous colony Queen number One Genotype of queen(s) BB heading colony Queen phenotype High fat reserves, rapid oogenesis (advantageous under monogyny) Worker phenotype BB workers in absence of Bb workers are intolerant of multiple queens
aSee
Polygynous colony Many Bb Low fat reserves, slow oogenesis
Bb workers tolerate multiple queens, but only if they bear the b allele; Bb workers detect and kill BB queens; all bb queens (and workers) die early because b is a recessive lethal
for example Refs [7,8,11].
0168-9525/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S0168-9525(02)02655-0
222
Research Update
Sequencing the Gp-9 gene
Polymorphism in Gp-9 was first detected by protein electrophoresis. By separating the Gp-9 protein electrophoretically and designing degenerate primers on the basis of its amino acid sequence, Krieger and Ross [4] have now isolated and sequenced the Gp-9 gene. The gene turns out to be 1700 bp and to contain five exons and four introns. GenBank BLASTX searches showed its closest match to be genes encoding pheromone-binding proteins in moths. Because these proteins are used in the chemical recognition of conspecific individuals, this is satisfyingly consistent with the presumed effect of Gp-9 in allowing ‘green beard’ recognition of allelic co-bearers. By sequencing the Gp-9 gene from queens of both social forms, Krieger and Ross also determined the nature of allelic variation in the gene: the two alleles differ by nine nucleotide substitutions, eight of which are nonsynonymous; that is, leading to amino acid differences in the alleles’ protein products [4]. Phylogenetics of Gp-9
In its native South America, S. invicta also occurs in polygynous populations, but they do not appear to be unicolonial [15]. Krieger and Ross [4] found that S. invicta queens from polygynous colonies from Argentina always carried a b allele, although in two variants showing slight sequence differences from the b allele in US populations. By sequencing Gp-9 loci in nine additional Solenopsis species and reconstructing their phylogeny, the investigators found a remarkable pattern: obligately monogynous species in the fire ant group possess only B-like alleles and are basal (ancestral to the other groups), whereas nearly all species known to be socially polymorphic had b-like alleles in the polygynous colonies. So, Krieger and Ross suggest that the b allele arose in an ancestrally monogynous BB population and has persisted during subsequent, multiple speciation events, always retaining its influence on social phenotype. Analysis of the ratio of nonsynonymous to synonymous nucleotide substitutions within lineages indicated that the divergence of b-like alleles from B-like alleles has been driven repeatedly by positive selection. Conclusions and prospects
A major achievement of this work is, as Krieger and Ross [4] stress, its http://tig.trends.com
TRENDS in Genetics Vol.18 No.5 May 2002
confirmation that a single gene with a large effect can underlie complex social behaviour. This is contrary to the intuitive view that complex behaviour is influenced by many genes with small effects (but see below). Whether the involvement of genes with major effects will prove exceptional or common in social evolution remains unclear (c.f. Ref. [16]). Another key achievement is pioneering the integrated study of social evolution across all levels of the biological hierarchy. We are now much closer to a complete understanding of a gene for social behaviour, from its sequence identity and protein product, through to its subsequent effects on individual behaviour, through to the summed outcome for the colony’s social structure and the gene’s fitness under natural selection. For this reason, Krieger and Ross’s work marks a crucial stage in the maturation of the study of social evolution as a scientific discipline. In future, it will be important to achieve a similarly comprehensive understanding of comparable genes in other social contexts and taxa. In this respect, note that Gp-9 is probably unusual because it is not a strictly kinselected gene (workers with the b allele favour b-bearing queens irrespective of kinship), and because it has remained polymorphic through b being recessive lethal [13]. The challenge remains to subject kin-selected genes to investigations such as that of Krieger and Ross. This could involve examining facultatively social Hymenoptera, whose populations contain both solitary and social nests (e.g. Ref. [17]). In principle, a single-gene difference could underlie the difference between these two social phenotypes (e.g. presence or absence of a gene for offspring dispersal [2]). However, this phenotypic difference could also stem from conditional expression of alreadyfixed alleles [18]. In theory, a kin-selected gene for helping behaviour retains conditional expression of helper and non-helper phenotypes even at fixation (e.g. Ref. [19]). The next comparable studies could, therefore, be ones where, as in S. invicta, genetic polymorphism is maintained by special circumstances, or conditional expression seems unlikely, or both. A candidate is the heritable ‘anarchic syndrome’ in honey bees, whereby a rare subset of colonies with queens routinely contain reproductive workers [20]. However, researchers will
rarely have the head-start of knowing that the expression of a social trait is correlated with a protein polymorphism. Other remaining questions concern the evolution of polygyny in S. invicta. Why is the b allele associated with multicolonial polygyny in South America but unicolonial polygyny in North America? Why, given the b allele’s presence in South American populations, did polygyny take 40 years to appear in the USA? Does the b allele bring about polygyny as a social novelty, or does it secondarily invade polygynous systems present for ecological reasons? In establishing the feasibility of tackling all these issues, the superb study of Krieger and Ross [4] has set a new standard in genetic research on social evolution. Acknowledgements
I thank Michael Krieger and Tracey Chapman for helpful comments. References 1 Hamilton, W.D. (1963) The evolution of altruistic behavior. Am. Nat. 97, 354–356 2 Bourke, A.F.G. and Franks, N.R. (1995) Social Evolution in Ants, Princeton University Press 3 Crozier, R.H. and Pamilo, P. (1996) Evolution of Social Insect Colonies: Sex Allocation and Kin Selection, Oxford University Press 4 Krieger, M.J.B. and Ross, K.G. (2002) Identification of a major gene regulating complex social behavior. Science 295, 328–332 5 Ross, K.G. and Keller, L. (1995) Ecology and evolution of social organization: insights from fire ants and other highly eusocial insects. Annu. Rev. Ecol. Syst. 26, 631–656 6 Chapman, R.E. and Bourke, A.F.G. (2001) The influence of sociality on the conservation biology of social insects. Ecol. Lett. 4, 650–662 7 Ross, K.G. (1992) Strong selection on a gene that influences reproductive competition in a social insect. Nature 355, 347–349 8 Keller, L. and Ross, K.G. (1993) Phenotypic basis of reproductive success in a social insect: genetic and social determinants. Science 260, 1107–1110 9 Ross, K.G. et al. (1996) Simple genetic basis for important social traits in the fire ant Solenopsis invicta. Evolution 50, 2387–2399 10 Ross, K.G. (1997) Multilocus evolution in fire ants: effects of selection, gene flow and recombination. Genetics 145, 961–974 11 Ross, K.G. and Keller, L. (1998) Genetic control of social organization in an ant. Proc. Natl. Acad. Sci. U. S. A. 95, 14232–14237 12 Keller, L. and Ross, K.G. (1999) Major gene effects on phenotype and fitness: the relative roles of Pgm-3 and Gp-9 in introduced populations of the fire ant Solenopsis invicta. J. Evol. Biol. 12, 672–680 13 Keller, L. and Ross, K.G. (1998) Selfish genes: a green beard in the red fire ant. Nature 394, 573–575 14 Dawkins, R. (1976) The Selfish Gene, Oxford University Press
Research Update
15 Ross, K.G. et al. (1996) Social evolution in a new environment: the case of introduced fire ants. Proc. Natl. Acad. Sci. U. S. A. 93, 3021–3025 16 Orr, H.A. and Coyne, J.A. (1992) The genetics of adaptation: a reassessment. Am. Nat. 140, 725–742 17 Michener, C.D. (1985) From solitary to eusocial: need there be a series of intervening
TRENDS in Genetics Vol.18 No.5 May 2002
species? In Experimental Behavioral Ecology and Sociobiology (Hölldobler, B. and Lindauer, M., eds), pp. 293–305, Gustav Fischer Verlag 18 West-Eberhard, M.J. (1989) Phenotypic plasticity and the origins of diversity. Annu. Rev. Ecol. Syst. 20, 249–278 19 Parker, G.A. (1989) Hamilton’s rule and conditionality. Ethol. Ecol. Evol. 1, 195–211
223
20 Barron, A.B. et al. (2001) Worker reproduction in honey-bees (Apis) and the anarchic syndrome: a review. Behav. Ecol. Sociobiol. 50, 199–208
Andrew F.G. Bourke Institute of Zoology, Zoological Society of London, Regent’s Park, London, UK NW1 4RY e-mail: [email protected]
Antisense RNAs everywhere? E. Gerhart H. Wagner and Klas Flärdh In recent years, systematic searches of both prokaryote and eukaryote genomes have identified a staggering number of small RNAs, the biological functions of which remain unknown. Small RNA-based regulators are well known from bacterial plasmids. They act on target RNAs by sequence complementarity; that is, they are antisense RNAs. Recent findings suggest that many of the novel orphan RNAs encoded by bacterial and eukaryotic chromosomes might also belong to a ubiquitous, heterogeneous class of antisense regulators of gene expression.
Antisense RNAs can be powerful tools to downregulate, at the post-transcriptional level, the expression of targeted genes. In a recent report, a conceptually simple antisense approach was shown to work remarkably well in the Gram-positive bacterium Staphyloccocus aureus [1]. These authors shotgun-cloned genomic segments and generated antisense RNAs from an inducible promoter on the plasmid. Many transformants displayed lethal phenotypes, or severe growth retardation, upon induction. This identified many essential genes that were previously unknown. In Escherichia coli, artificial antisense RNA strategies have been used for many years [2]. Unfortunately, the efficiency of introduced antisense RNA genes in these Gram-negative bacteria varies greatly, and attempts to improve the efficiency by rational design have been rather disappointing [3]. Currently, specific antisense RNA efficiencies cannot be compared between bacterial systems, because intracellular concentrations have generally not been determined. Nevertheless, S. aureus appears to be superior to E. coli in supporting antisense RNA-mediated downregulation of gene expression. Whether this is owing to http://tig.trends.com
differences in intracellular conditions or the presence or absence of certain enzymatic activities is as yet unkown. Antisense RNA and co-suppression (sense RNA) have also been used successfully in plants. Current wisdom suggests that both approaches are related to the natural process of dsRNA-mediated interference (RNAi) [4]. In light of the justified excitement over the successful use of antisense RNA-based approaches, it is surprising that natural antisense RNAs receive little attention. After all, these RNAs do it for a living and might tell us about optimization principles. Antisense interactions involved in biological functions
In Nature, many different biological activities work through an antisense principle. In eukaryotes, processes such as splicing, RNA editing, rRNA modification, and developmental regulation rely on base-pairing between complementary RNAs (or stretches thereof). Small RNAs are the key elements: spliceosomal snRNAs, gRNAs, snoRNAs and stRNAs (see Glossary for definition of RNAs). The recent discovery of RNAi, initially established in Caenorhabditis elegans,
adds to this list [5]. In RNAi, dsRNA is cleaved by the enzyme Dicer, into ~21-nucleotide (nt) dsRNAs called small interfering RNAs (siRNAs). Mediated by an enzyme complex, RISC, target RNA is bound by the complementary RNA – one of the two strands in the short duplex – and subsequently degraded [4]. The antisense principle is crucial for the specificity of target recognition and silencing, and in this context presumably acts to defend against viruses and transposition events. Antisense RNAs as regulators
Historically, Jacob and Monod [6] first suggested that gene regulation in bacteria could be carried out by RNAs (or proteins). About 20 years later, small untranslated antisense RNAs were discovered in bacterial plasmids and shown to regulate the copy numbers of two plasmids, ColE1 [7] and R1 [8]. RNAI (of ColE1; 108 nt) and CopA (of R1; 93 nt) both bind rapidly and irreversibly to their respective target RNAs to block primer RNA maturation or inhibit Rep protein translation, respectively (Fig. 1). Regulation occurs by negative feedback. Numerous studies have given a detailed understanding of both the biology and biochemistry of these systems (for reviews, see Refs [9–11]).
Glossary of RNA terms dsRNA: Double-stranded RNA. gRNA: Short guide RNAs that bind to complementary pre-mRNAs to specify editing. miRNA: Micro RNAs, a class of 22- to 25-nucleotide short RNAs in animals. Functions are as yet unknown, but might be stRNAs. ncRNA: Noncoding RNA. In general, these are non-messenger RNAs. Riboregulator: Regulatory RNA (term used in bacteria). RNAi: dsRNA-mediated RNA interference. Gene silencing using introduced dsRNA with homology to the target RNA. sRNA: Small noncoding RNA. A class of putative riboregulators. siRNA: Short, double-stranded ~22mers that are generated during RNAi; siRNAs mediate subsequent target RNA degradation. snRNA: Heterogeneous class of small nuclear RNAs; for example those involved in splicing. snoRNA: Small nucleolar RNA. Directs rRNA modification. stRNA: Small temporal RNA such as lin-4 and let-7, involved in developmental control.
0168-9525/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S0168-9525(02)02658-6